2022

DATA ANALYSIS
[University of Roehampton Business School]
Contents
1. Introduction.........................................................................................................................................3
2.0. Analysis of shopping data............................................................................................................3
2.1.1. Insight 1: Descriptive Analysis..............................................................................................3
2.1.2. Question:.............................................................................................................................3
2.1.3. Method................................................................................................................................4
2.1.4. Results.................................................................................................................................4
2.1.5. Recommendations...............................................................................................................4
2.2. Insight 2: Device Usage............................................................................................................4
2.2.1. Question:.............................................................................................................................4
2.2.2. Method................................................................................................................................4
2.2.3. Results.................................................................................................................................5
2.2.4. Recommendations...............................................................................................................5
2.3. Insight 3: Bounce Rate.............................................................................................................5
2.3.1. Question:.............................................................................................................................5
2.3.2. Method................................................................................................................................5
2.3.3. Result...................................................................................................................................6
2.3.4. Recommendations...............................................................................................................6
2.4. Insight 4: Net Promoter Score (NPS)........................................................................................6
2.4.1. Question:.............................................................................................................................6
2.4.2. Method................................................................................................................................7
2.4.3. Result...................................................................................................................................7
2.4.4. Recommendation.................................................................................................................7
2.5. Insight 5. Email delivered Vs Email Subscribe..........................................................................7
2.5.1. Question:.............................................................................................................................7
2.5.2. Method................................................................................................................................7
2.5.3. Results.................................................................................................................................8
2.5.4. Recommendation.................................................................................................................8
3.0. 3.0. Impact of Data Analytics on Marketing Strategies................................................................9
3.1. Overview......................................................................................................................................9
3.2. Data Analytics..............................................................................................................................9
3.3. Actionable Insights.....................................................................................................................10
3.4. Role of Big Data Technology 5...................................................................................................11
3.4.1. Why is big data analytics important?.....................................................................................11
4.0. Conclusion.................................................................................................................................12
5.0. References.................................................................................................................................13
List of Figures Figure..............................................................................................................................14
1. Introduction
The purchaser is unquestionably the extremely imperative root in any business. You may learn
the channels your customers use to connect with you, their characteristics, hobbies, habits,
purchasing patterns, and more by using analytics to observe all elements of your consumers
360 degrees (Calzon, 2022). In the long run, this can help your marketing strategy succeed,
identify new potential customers, and avoid wasting resources by targeting the wrong person
or sending the wrong message. You can also track customer satisfaction by analyzing customer
reviews and customer service department performance.
This can help your marketing plan succeed in the long term by identifying new prospective
consumers and avoiding wasting money by targeting the wrong individual or communicating
the incorrect message. Customer satisfaction may also be tracked by monitoring customer
feedback and customer relations department performance. Analyzing your data may help you
make factual judgments rather than intuitive ones, which is beneficial from a management
standpoint.
For instance, you can figure out where to put your money, spot growth prospects, forecast
income, and deal with odd events before they become an issue. You may use dashboard
software to deliver information to many participants in a competent and engaging manner by
extracting essential insights from all sections of your firm. Another significant benefit is the cost
savings. With the assistance of. Over time, this will help save money and resources in
implementing the wrong strategy. In addition, you can predict production and supply by
forecasting different scenarios such as sales and demand.

2.0. Analysis of shopping data

Now, if we look into the data set of shopping which is from the "fashion website in 2020". We
see different data points. To analyze all the points, I’ll use different techniques including
formulas when needed.
2.1.1. Insight 1: Descriptive Analysis
For finalizing and better understand our customers, we have to analyze the shopping
data with different types of techniques. Because this is the only way to get more useful
insights from the data. So, here we have applied Descriptive Analysis, which does not
help in future predictions but helps a lot in understanding the data.
The main reason behind this analysis is to better understand the data points such as
"Where does the customer come from? How many countries are included? and What
sort of marketing techniques are used?
 2.1.2. Question:
How many data points are in the provided shopping data set?
There are mainly eleven main data points to focus on. These data points are " Country
the customers are from, Device the customers use, age of the customer, gender, NPS
"on a scale of 10 how many consumers recommend their friends and family to visit the
website? Customer relationship, Customer acquisition cost, Annual Revenue per
customer, delivery of email, did the email marketing work, and did the consumer
subscribe to the newsletter? Along with the number of customers used for the analysis,
which is 204.

2.1.3. Method
Using descriptive analysis, I have gone through the data given in the excel sheet.

2.1.4. Results
Found these columns in the sheet and further analysis is going to be on the columns.

Data set

2.1.5. Recommendations
This data is not enough to analyze the nature of the customer against any data point.
More survives and online interviews should be done for better results.

2.2. Insight 2: Device Usage

In this era, we not only have the systems and laptops but we do use Mobile phones and
Tablets. We found out the Dataset of shopping contains a column named “Device”
which has the names of the devices the consumers used while shopping and scrolling
the website. This information is really important because this helps to understand more
about the nature of our consumers. Based on this information, it is possible to identify
easily what device is being used more frequently.
2.2.1. Question:
How many devices are used and What are the names of the devices?
According to the information provided in the Data set shopping the number of devices
used is four (4). Mobile Phone, Laptop, Desktop, and Tablet.
 2.2.2. Method
The method here used is quantitative analysis. Performing quantities analysis is
significant, it turns the data into information and gives insights in a meaningful way.
Thus, it this technique does not provide any future predictions, like why something
happened, it gives really meaningful insights.
2.2.3. Results
The shopping data contains 204 devices altogether. Where the number of times
Mobile Phones were used by the consumers is (133). The number of times Laptops
were used is (14). The number of times Desktops were used is (34). And the number of
times Tablets were used is (23). So, we can clearly see that the most frequently used
device by the consumers is the Mobile Phone.

2.2.4. Recommendations
The results show that the website is used more through the Mobile Phones, which is a
good point. It shows that the application is mobile-friendly. So, we recommend
checking if the website runs okay on the other devices, if yes, this means the users
prefer Mobile more than any other device, but if no, then we definitely need to make
the Shopping website better.

2.3. Insight 3: Bounce Rate

The bounce rate shows the time between a visitor’s visit and stays on the website.
Knowing the bounce rate is important because it gives really meaningful insights and
we can predict a lot of insights by just the bounce rate.

2.3.1. Question:
What does the data in the bounce rate column show?
Based on the data declared in the data set there are very few bounced. This means the
consumers came on the website and left the website within a minute without
spending time on the website.
 2.3.2. Method
Bounce= Total number of one-page visits / Total number of entries to a website
(WP, 2022)

2.3.3. Result

Bounce Rate

35%

65%

Bounced Not Bounced

Bounce Rate
According to the data mentioned in the shopping data set, 36 out of 104 bounced.
Which means 34.6% bounced.

2.3.4. Recommendations
25% to 41% bounce rate considered as the best bounce rate, but if we want to improve
it more. We should start ads that are more pleasing to the customers.
2.4. Insight 4: Net Promoter Score (NPS)
The Net Promoter Score (Rbucklin, n.d.) is indeed a statistic that measures how willing
customers are to suggest a company and its products to others. It runs from -100 to 100.
It is also used to analyze the total degree of consumer contentment with a company’s
products and services, as well as customer brand loyalty. (Medallia, 2010 ).The results
below show the different loyalty levels in different countries. Question asked to check
the number of loyal people. Customers in each country provide a location-based view of
loyalty in different countries' geographic markets. Based on this information, it is
possible to identify the countries where their customers are more loyal and identify
customer segments that may need further development.
2.4.1. Question:
Analyze the persistent customers according to their location?
To analyze the loyal customers from a country we have to focus on the NPS column.
Further, the column is divided into two parts. The Scores from (0-6) and the scores from
(9-10).
 2.4.2. Method
The column contains numbers in the range of 10. For the specific research the included
countries targeted are USA, UK, Sweden, along with Brazil. The following formula was used to
compute the net promoter of each kingdom's score:
NPS = Percentage of Promoters (%) – Percentage of Detractors (%) (CFI, 2022)

2.4.3. Result
The table below is showing the number of promoters, number of detractors in percentage, and
the overall NPS score by the intended region. Inspecting the data in the table. It is possible to
see Brazil has the uppermost NPS score (55.7), indicating the most committed customers.

Country Percentage of Promoters Percentage of Detractors NSP = Percentage of Promoters

(%) – Percentage of Detractors
(%)
US 1.4 % 11.27 % -9.87
UK 1.9 % 38.4 % -36.5
Brazil 0% 55.7 % 55.7
Sweden 2.88 % 34.6 % -31.72

2.4.4. Recommendation
Derived from the results it is advisable to consider in depth examine why NPS scores are
miserably down and negative in US, UK, and Sweden.

2.5. Insight 5. Email delivered Vs Email Subscribe

When a customer subscribes for email, means they are interested in our content/products or
website. Here we’ll look deeply into and try to find out if we have got email subscribers and did
that make any transaction.

2.5.1. Question:
Did the customer subscribed after receiving email?
To dig more into the data, we’ll analyze both the columns ‘Email delivers and Email
subscribed”
2.5.2. Method
Exploratory Data Analysis method is being used here. So, we can easily look into both
of the columns each data.
 2.5.3. Results

Email Delivered VS subscribed

74

68

36
30

Delivered Not Delivered Suscribed Not Subscribed

According to the results out of 104 delivered emails 68 email delivered and 36 didn’t
deliver. On the other hand, just 30 of the customers subscribed for newsletter. Which
means we got 28.8% of email subscribers from the 4 countries.

2.5.4. Recommendation
The data needs to be explored more specifically, so we can find out what country is
producing more subscribers. After knowing the targeted market, we can do further
marketing on that region which can produce more results.
3.0. Impact of Data Analytics on Marketing Strategies
3.1. Overview
E-commerce serves as a connecting point for customers and merchants. Data analytics is
used by companies to measure the great sense of products as compared to other
suppliers. Organizations may enhance their study of transaction trends, service and
support, forecasting future strategic plan, segments and sub attention, and online
banking convenience.
People constantly use the Online services to exchange information and purchase and
sell goods. E-commerce businesses have utilized the Web to offer goods and services,
boost profitability, and raise brand awareness. Consumers could only buy products or
items online by visiting the website now that online purchasing has become a craze.
Online shopping allows customers to buy swiftly and save time. In the modern e-
commerce system, managing a huge number of items and consumers is tough. The
number of items maintained on a particular online shopping website is restricted, and
the commodities are less likely to be recommended in the offered recommendation
system. Given these constraints, a method for moving resources across 2 or 3 servers to
just one is required.
Social media platforms are linked not just to individuals’ networks, but also to virtual
communities where people communicate, share information, and express their ideas.
Predictive analysis analyses future trends based on previous and longitudinal studies. To
find trends and learn from past data, this study uses mathematical tools and a machine-
learning algorithm. Big data analysis is employed in e-commerce, social networks, and
telecom networks, among other places. Hadoop frameworks are therefore controlled
through using map reduction concept, which enhances product sustainability across a
wide range of products and handles a big number of customers.

3.2. Data Analytics

Data analytics is the study of how data is transformed into important knowledge. After
the data is analyzed, the companies may enhance purchase patterns analytics, effective
customer service, anticipate future operations, increase attention on micro-moments,
and make digital purchases easier by employing data analytics. Humans increasingly use
the Internet to communicate content and purchase and sell goods. E-commerce
businesses are already leveraging the Internet to trade goods and services, boost
income, and build brand awareness (Moorthi, 2020).
A methodology demonstrating the impact of social media was constructed by combining
data from traditional media, sales, and consumer usage from major social networking
sites. Much better results than the company's strategists had predicted. More crucially,
the company's analytics discovered that consumer engagement with the website, not
email advertising, was the key generator of sentiment on the social network. Company
has dramatically boosted its client base and made millions of dollars in sales by
reallocating part of its content budget to strengthen emails' CTA. So much data may be
overwhelming.
That is why it is critical to make the remarks straightforward. For example, a company
may recognize that a significant portion of its sales originate from a limited number of
consumers, but that growth in revenue to those customers is moderate. Local sales staff
will be expected to locate new consumers, according to officials. As a result, the firm
established a central analytics team that gathers thorough data and develops predictive
models that allow the company to discover local marketplaces with the best sales
potential for new clients. Rather of overloading sales staff with data and sophisticated
algorithms, the team developed a powerful tool with a user-friendly interface that aids
in the identification of possible new consumers by zip code.
This application enables sales managers to proceed to the county level and evaluate zip
codes with high development potential, then deploy their sales teams to those areas.
The usage of this instrument had helped the firm to quadruple its sales growth while
successfully lowering its cost of sales by the time it was completed (Goyal, 2015).

3.3. Actionable Insights

Much of CRM revolves not only around collecting information about the customer, but
also around information that explains the customer. This information can provide
insights into future customer behaviors, needs, and actions, so businesses need to know
how to extract these insights from customer feedback.
Ideally, these insights are actionable indicators pointing to improvements in sales and
marketing efforts to improve the relationship between the customer and the company,
and ultimately improve the customer experience. (Ronald T. Wilcox, 2014) Indicators
include securing a sale, increasing brand loyalty, improving service, or converting a lead
into a customer. For indicators to be considered actionable, they must be verifiable.
The important part of this is that companies need to understand that not all insights are
actionable, so just leave them aside. Furthermore, data is only insightful when it is about
something a company did not know or could not learn from already established sources.
Ultimately, actionable insights can be an effective barometer of how well a campaign or
customer-centric process is performing, and businesses should use them as such.
(Ronald T. Wilcox, 2014)
3.4. Role of Big Data Technology
Big data analytics studies vast volumes of data to identify hidden information, linkages,
and new insights. Through present tech, you can study your data and acquire answers
from it right away – a process that is longer and less cost-effective previous traditional
enterprise systems (Wiley, 2013).
The way data is created, handled, analyzed, and utilized is fundamentally altering
because to technological advancements. On the one hand, technology aids the
development of new and more effective sources of data. On the other hand, when more
data is collected, innovation comes in to assist process it quickly and effectively, as well
as display it so that educated decisions may be made. More than at any other moment
in the brief history of analytics, technology is becoming increasingly important in the
complete conduct of this study and analysis. (Wiley, 2013).
Big Data is a developing sector, with new technology and terminology developing on a
fairly regular basis (Wiley, 2013). The accessible ecosystem, rising technological
businesses, and corporate Armored knights like IBM, Oracle, SAP, SAS, and Teradata all
contribute to this fast-paced environment.

3.4.1. Why is big data analytics important?

Reduce costs: The Big data is providing immense types of facilities in the field of
analytics and marketing. Reducing the costs of marketing is it’s one of the best
features. As, it digs deep and brings out the exact information that businesses
should focus on.
Make better, quicker conclusions: To run the business at higher level with a
good speed, we need to quickly analyze consumers and make decisions. This is
where big data analytics helps out, because it is the fastest technology with
widest techniques.
Development and marketing of new products and services: It allows us to
understand the buyers interest and need. By using the analytics through the big
data companies and businesses can figure out and launch new products
accordingly.
Right Audience: We can find out from where our right audience is coming from
and we can focus on them. For targeting other audiences, we can do test
research on other marketing strategies. After getting the positive results from
any specific marketing strategy we can further go with the strategy to get more
result in better way. This way, we can focus on two things at a time and can
analyze the results for further improvement.
 4.0. Conclusion
Going through each aspect of marketing and data analytics I’m concluding that managers ought
to recognize their advertising efforts as exactly as viable to decide how a lot to spend on every
advertising channel. If paid marketing working and required then paid marketing is the simplest
manner of having a firm’s message in front of the proper client, why might the organization
spend extra on print marketing and marketing? If income calls are worthwhile and most
effective as much as a factor, the advertising supervisor ought to recognize at which factor the
calls begin costing his or her organization cash in place of making it.

The most effective manner to degree the results of advertising efforts on profitability is thru the
best-bet relationships discovered thru advertising analytics. By the usage of statistical evaluation
techniques, corporations can use beyond client behaviors to are expecting how clients will react
to distinct advertising channels; managers can then optimize spending on every channel.
References

Calzon, B., 2022. Your Modern Business Guide To Data Analysis Methods And Techniques. Data Analysis.

CFI, 2022. Net Promoter Score (NPS). 02 Feb.

Goyal, M., 2015. Marketing and sales. Big data analytics, Future sale, p. 60.

institute, C. c. f., n.d. Net Promoter Score (NPS).

Medallia, 2010 . Net Promoter Score Defination.

Moorthi, K., 2020. A survey on impact of data analytics techniques in E-commerce. ScienceDirect, p. 8.

Rbucklin, n.d. NET PROMOTER SCORE – NPS. Insight Survey .

Ronald T. Wilcox, R. V. P. W. F., 2014. Cutting Edge Marketing Analytics: Real World Cases and Data Sets
for Hands On Learning (FT Press Analytics). First ed. s.l.:s.n.

Ronald T. Wilcox, R. V. P. W. F., 2014. Cutting- edge Marketing Analytics. First ed. s.l.:s.n.

Wiley, 2013. Big Data, Big Analytics: Emerging Business Intelligence and Analytic Trends for Today's
Businesses. First ed. s.l.:s.n.

WP, S., 2022. 17 Proven Tactics to Reduce Your Bounce Rate and Increase Conversions. 29 March.
List of Figures Figure
Figure 1: Data set

Figure 2: Data set

Figure 3: Net Promoter Score by country

Figure 4: Email delivered Vs Email Subscribe